Fig. 8: Evolutionary dynamics of neuroblastoma initiation.
From: Neuroblastoma arises in early fetal development and its evolutionary duration predicts outcome
a, Assuming a constant putative population of origin following neuroblast expansion (left panel), the model fit does not capture the observed saturating incidence of high-risk neuroblastomas (right panel; shaded areas, 95% posterior probability of the fit; data, mutation densities of the MRCA from primary neuroblastomas with an acquired TMM, combining discovery and validation cohort; solid line, maximum-likelihood estimates of data with error bars representing s.d. estimated by bootstrapping). b, Transient putative population of origin. c,d, Model fits to ECA (c) and MRCA (d) with transient population of origin, as in b accounts for the experimental data [green shaded areas, 95% posterior probability bounds; vertical lines and shaded areas, mean and 95% CI of the estimated end of the first trimester (12 weeks p.c.) and of the time of birth (38 weeks p.c.); data, mutation densities of ECA (dark green, n = 47) and MRCA (light green, n = 95) from primary neuroblastomas (tumors/metastases) with acquired TMM; solid lines, maximum-likelihood estimates; error bars, s.d. estimated by bootstrapping]. e, Predicted transient expansion of putative cells of origin, agreeing with rapid proliferative phase of sympathetic neuroblasts (shaded area, 95% posterior probability; vertical line and shaded area, mean and 95% CI of the estimated end of the first trimester (12 weeks p.c.)). f, Estimated mutation rate per effective cell division, computed from primary tumors/metastases with MYCN amplification (amp.) (n = 11), TERT rearrangement (n = 2), ALT (n = 14) or no acquired TMMs (n = 15), using cases with highly accurate subclonal VAF distribution due to high tumor purity. Boxes represent median and 25 and 75% percentiles; whiskers extend to the smallest and largest values within 1.5× interquartile range. g,h, Estimated loss rate (relative to cell division, g) and cell division rate (h) in primary tumors and metastases analyzed in f (shown are n = 11 primary tumors with MYCN amplification, n = 2 primary tumors with TERT rearrangement, n = 14 primary tumors with ALT and n = 15 primary tumors without acquired telomere maintenance). Boxes represent median and 25 and 75% percentiles; whiskers extend to the smallest and largest values within 1.5× interquartile range; λT, division rate during tumor growth; δT, loss rate during tumor growth.
